Michael A. Högele,Alexander Steinicke
Abstract:This case study proposes robustness quantifications of many classical sample path properties of Brownian motion in terms of the (mean) deviation frequencies along typical a.s.~approximations. This includes Lévy's construction of Brownian motion, the Kolmogorov-Chentsov (and the Kolmogorov-Totoki) continuity theorem, Lévy's modulus of continuity, the Paley-Wiener-Zygmund theorem, the a.s.~approximation of the quadratic variation as well as the laws of the iterated logarithm by Khinchin, Chung and Strassen, among others.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to quantify the robustness of Brownian motion path properties, especially by studying the deviation frequencies of these properties under typical almost - sure (a.s.) approximations. Specifically, the paper focuses on how to quantify the robustness of many classical sample - path properties through deviation frequencies, including:
1. **Lévy's Construction of Brownian Motion**: Study the almost - sure convergence rate of Brownian motion paths in Lévy's construction method and its deviation frequency.
2. **Kolmogorov - Chentsov Continuity Theorem**: Quantify the deviation frequency of the Hölder continuity of stochastic process paths.
3. **Lévy's Modulus of Continuity**: Study the modulus of continuity of Brownian motion paths and its deviation frequency.
4. **Paley - Wiener - Zygmund Theorem**: Quantify the quantitative divergence of secant slopes of Brownian motion paths.
5. **Quantitative Loss of Path Monotonicity**: Study the probability that Brownian motion paths do not have monotonicity in any sub - interval.
6. **Law of the Iterated Logarithm**: Study the deviation frequencies in the law of the iterated logarithm of Khinchin, Chung and Strassen.
### Main Contributions
1. **Statistical Quantification of Deviation Frequencies**:
- The paper proposes an intuitive and widely applicable statistical quantification method for measuring the random frequency of deviation from the limit value at each error level.
- Through this method, the author studies the almost - sure convergence rate of Brownian motion paths in Lévy's Haar - basis construction and gives the corresponding deviation frequency.
2. **Improved Classical Results**:
- In some cases, the concept of deviation frequency can provide more refined results, such as improving the L2 convergence rate of Lévy's construction.
- Through the analysis of deviation frequencies, the paper also provides an accurate quantitative distinction regarding the asymptotic behavior of Brownian motion near 0.
3. **Application of the Borel - Cantelli Lemma**:
- Using the moment estimates in the Borel - Cantelli lemma, the paper shows the trade - off between almost - sure error tolerance sequences and error - occurrence frequencies.
- This trade - off indicates that the faster the almost - sure error tolerance approaches 0, the slower the decay of the corresponding error - occurrence frequency (in a non - linear sense).
4. **Numerical Solutions and Applications**:
- The method of deviation frequencies can be applied to many other scenarios, such as numerical solutions of stochastic (partial) differential equations, Lévy and additive processes, etc.
- The paper shows how to implement these findings through multiple examples and annotations.
### Specific Results
- **Almost - Sure Convergence Rate of Lévy's Construction**:
- Theorem 1 gives detailed estimates of the almost - sure convergence rate of Brownian motion paths in Lévy's construction and its deviation frequency.
- The results include almost - sure random upper bounds, almost - sure deterministic upper bounds, step - by - step adapted error deviation frequencies, and deviation frequencies for fixed errors.
- **Kolmogorov - Chentsov - Type Results**:
- Theorem 3 and Theorem 4 respectively give the deviation - frequency quantifications of the Hölder continuous versions of stochastic processes and random fields.
- These results are applicable to Brownian motion, fractional Brownian motion, and Brownian sheets, etc.
- **Deviation Frequencies of Fine Continuity Properties**:
- Theorem 5 studies Lévy's modulus of continuity and gives a Gumbel - type decay of deviation frequencies.
- Theorem 6 and Theorem 7 respectively study the Paley - Wiener - Zygmund theorem and the loss of monotonicity of Brownian motion paths and provide specific estimates of deviation frequencies.
### Conclusion
By introducing the concept of deviation frequency, the paper provides a new method to quantify the robustness of Brownian motion path properties. This method not only improves some classical results but also provides a useful tool for numerical analysis and simulation.